model_276347518_coca_base.py

Model Overview

A base-scale implementation of the coca architecture, built for retrieval tasks.

Architecture

• Architecture: coca
• Scale: base
• Attention: flash
• Fusion strategy: low rank
• Task head: retrieval
• Activation: relu
• Normalization: scalenorm
• Initialization: trunc normal

Training

• Optimizer: adamw
• LR scheduler: onecycle

Files

• model_276347518_coca_base.py — main artifact of this repository

License

See the license field above.